DANCE
Domain Adaptative Neighborhood Clustering via Entropy Optimization
Description
Domain Adaptive Neighborhood Clustering via Entropy Optimization (DANCE) is a self-supervised clustering method that harnesses the cluster structure of the target domain using self-supervision. This is done with a neighborhood clustering technique that self-supervises feature learning in the target. At the same time, useful source features and class boundaries are preserved and adapted with a partial domain alignment loss that the authors refer to as entropy separation loss. This loss allows the model to either match each target example with the source, or reject it as unknown.
Papers Using This Method
Dataset Distillers Are Good Label Denoisers In the Wild2024-11-18DANCE: Deep Learning-Assisted Analysis of Protein Sequences Using Chaos Enhanced Kaleidoscopic Images2024-09-10Improving Commonsense in Vision-Language Models via Knowledge Graph Riddles2022-11-29Graph Neural Networks for Multimodal Single-Cell Data Integration2022-03-03Domain Adaptation of Networks for Camera Pose Estimation: Learning Camera Pose Estimation Without Pose Labels2021-11-29More Robust Dense Retrieval with Contrastive Dual Learning2021-07-16DANCE: DAta-Network Co-optimization for Efficient Segmentation Model Training and Inference2021-07-16Decorrelating Adversarial Nets for Clustering Mobile Network Data2021-03-11DANCE: Differentiable Accelerator/Network Co-Exploration2020-09-14Universal Domain Adaptation through Self Supervision2020-02-19